OpenAI says its Oct. 21 expansion adds ChatGPT Ads in 26 more countries, bringing availability to 89. The company says ads may appear for Free and Go users in supported markets, while several paid and organizational plans remain ad-free; the rollout does not mean every account or advertiser is eligible. OpenAI’s announcement and help pages describe the product and controls, while Reuters and CNBC independently reported the expansion; Reuters framed it as part of the company’s effort to grow advertising revenue.

What expanded on October 21

OpenAI announced the change on October 21, 2026: ChatGPT Ads would expand to 26 additional countries, taking availability to 89. The company said more than 100,000 advertisers had run campaigns since ads launched just over eight months earlier. Its announcement framed the service around commercial discovery: people use ChatGPT to explore ideas, compare options and make decisions, and a sponsored recommendation might introduce a product while a person is considering alternatives.

The timing also gives the expansion continued relevance despite being six days old. It is not just a dated launch notice: the expansion is part of an active, staged international rollout, and the conditions under which users encounter ads and advertisers enter the platform are still evolving. OpenAI says the next phase will include more formats, campaign objectives, buying options and measurement capabilities. Those planned changes make the initial country expansion a live commercial development rather than a completed one-off.

A 26-country rollout with a partial public list

The announcement says 14 of the new countries are in Latin America, following earlier launches in Mexico and Brazil. It names Argentina, Chile, Colombia and Peru as examples. OpenAI also identifies Nigeria and South Africa in Africa, Serbia and Ukraine in Europe, and Cambodia, Bangladesh and Pakistan in Asia. The post describes expansion across Europe, the Middle East, Africa and Asia, but does not enumerate every one of the 26 additions in the announcement itself. OpenAI’s October 21 announcement is therefore the primary source for the total and named examples, not a complete itemized list.

That distinction matters for companies deciding whether they can buy ads, and for readers trying to interpret “available in 89 countries.” OpenAI’s current Ads Manager availability page is a list of countries where self-service advertiser access is available. It is not necessarily a verbatim reconstruction of the 26-market consumer rollout: the company says advertisers can use technology partners or its Ads Manager, and the account route can affect access. Readers should check the current list instead of inferring every country from the headline total.

For advertisers using self-service, OpenAI says the legal entity that will advertise and be billed must be based in a listed country. Signing up in an unsupported country does not itself grant campaign access; the company may notify an account when access opens. Larger advertisers can approach OpenAI or partners for supported onboarding. In other words, consumer reach, a market’s inclusion in an expansion, and a particular advertiser’s ability to activate a campaign are related but not identical things. OpenAI’s availability guidance spells out that distinction.

Who can see ads—and who can buy them

On the user side, plan and age matter

OpenAI says ads may appear for users on its Free and Go plans. The October expansion announcement says Plus, Pro and Enterprise remain ad-free; its updated help material also names Business and Edu among ad-free accounts. That broader help-center list is the more complete current account guide. The company says it does not show ads to accounts identified as belonging to people under 18, using account age information and age prediction where available. Temporary Chats do not show ads, according to the help center.

On the advertiser side, country access is only the first gate

OpenAI’s rollout names technology partners and self-service Ads Manager as routes into the newly supported markets. A new self-service account may initially be limited to its home country. The current campaign guidance says wider international targeting can become available after identity verification is approved and the advertiser reaches a required level of spend from campaigns in the home market. Choosing another country in the campaign interface does not override that restriction. This means nominal platform availability does not guarantee immediate cross-border buying privileges.

There are also content rules. OpenAI’s ad policy permits consumer categories such as lifestyle and household goods, local services, travel, experiences, digital products and education in its initial test, while some sensitive or regulated verticals require case-by-case approval or remain disallowed. Ads must satisfy the policy across the creative and landing page. Political ads are disallowed, and ads cannot be placed near personal health, mental-health or political conversations. Advertiser access is thus broader geographically but still conditional on identity, billing, targeting and content review.

Separate placements, not sponsored answers

OpenAI says the ads do not appear as paid text woven into ChatGPT’s answer. The company says they are clearly labeled, displayed separately from the answer, and do not influence the response. Its current help center describes a placement below a response and a multi-image visual format being tested during image generation. The latter is a distinct experimental format; it does not mean that advertisers can rewrite or buy space inside the model’s generated answer.

The key distinction is between relevance and influence. OpenAI says it uses conversation context to select relevant ads, while denying that advertisers can shape or rank model responses. A user might discuss meal planning and see a sponsored grocery service below an answer. That is contextual advertising adjacent to a conversation, not an answer that has been purchased. The boundary is easy to state, but it must be maintained consistently in interface design and system behavior as the company adds formats. OpenAI says future formats and more native business interactions are under development, not that they are already standard placements.

Personalization, controls and sensitive conversations

ChatGPT’s ad controls let eligible Free and Go users turn off personalization, dismiss an ad, report it, see why it appeared, and clear data used for ads. Clearing data removes it from ad selection, with OpenAI’s help page saying deletion can take up to 30 days from its servers. Users who disable personalization may still see ads based on the current chat thread; the setting stops using other chat threads, ad history or topics to inform selection. Temporary Chat is described as an ad-free experience.

Personalization can use more than the current question when enabled. OpenAI’s help center says select signals from a user’s broader ChatGPT experience, including past chats and memory, may be used. That creates a tension the company has to manage: context may improve relevance, but users may reasonably find cross-conversation ad selection more sensitive than a placement based only on the current topic. The controls matter because they give users a way to limit that signal use, rather than pretending contextual selection is wholly detached from personal experience.

OpenAI says advertisers do not get access to chats, chat history, memories or personal details and do not receive users’ data for sale. Its current FAQ says reporting provided to advertisers is aggregated, such as total views or clicks. It also says no ads are eligible near sensitive topics, including personal health and mental health, and political content. These are company statements about system design and policy; they are not an independent audit of each serving decision. The current ads FAQ documents both the controls and the stated data boundaries.

There is a way for people on Free to avoid ads without moving to a paid subscription: OpenAI says an Ads-Free option is available with lower usage limits and reduced access to some tools. Plus and Pro offer paid ad-free options. That choice makes the access argument more concrete: advertising can subsidize a no-cost tier, but the no-ad alternative for a free user comes with a product trade-off. Availability of controls and options can vary by region and over time.

The business case and the proof still missing

Reuters reported that OpenAI said its advertising business had reached a $1 billion annualized revenue run rate in August. CNBC likewise reported the August figure and said it was up from $100 million in March. An annualized run rate is a snapshot extrapolated from the pace at a point in time, not revenue already earned across a full year. Reuters framed the international expansion in the context of OpenAI’s effort to bolster revenue ahead of a potential IPO. That is market context, not proof that the expansion itself was caused by a specific financing timetable.

The company’s 100,000-advertiser figure is another scale signal, but it is not a measure of how many advertisers are active now, how much they spend, or whether their campaigns produce profitable results. OpenAI says it plans to build new buying and measurement capabilities. Those plans point to an early-stage advertising business still developing its commercial infrastructure, not a mature network with publicly disclosed performance benchmarks for each market.

OpenAI’s own stated principles prioritize trust and user experience over revenue and say it does not optimize for time spent. The expansion therefore puts those principles under real operating pressure: more markets and advertisers can make the product more commercially useful, but also raise the stakes of ad relevance, policy enforcement and clarity. The relevant test is not simply whether the ad business grows. It is whether growth remains compatible with answers users can distinguish from paid messages and with privacy choices people understand.

What the rollout signals next

The October announcement is explicit about the next ambitions: more formats, objectives, buying options and measurement, as well as further exploration of ways for businesses to interact with consumers more natively in ChatGPT. The wording matters. These are development goals, not a promise that every planned format is available in all 89 countries. OpenAI’s August update described expanding the pilot carefully and learning from feedback, while its current help materials show controls and formats changing over time.

FAQs

How many countries are included after the expansion?

OpenAI said the October 21 expansion added 26 countries and brought ChatGPT Ads availability to 89. The announcement gives examples but does not list all 26 new countries.

Which ChatGPT users may see ads?

OpenAI says ads may appear on Free and Go plans in supported regions. Plus, Pro, Business, Enterprise and Edu accounts are ad-free, and accounts identified as belonging to users under 18 are excluded.

Are ads part of ChatGPT’s answers?

No. OpenAI says ads are clearly labeled and separate from answers. Its help center describes ads below responses and a visual format being tested during image generation.

Can users opt out of personalized ads?

Yes. Eligible users can switch off ad personalization, dismiss or report ads, see why an ad was shown, and clear ad data. Free users can also choose an ad-free option with lower limits and reduced feature access.

Can any advertiser target all 89 countries?

No. Self-service eligibility depends on the billing entity being based in a supported country, and some new accounts initially face home-country-only targeting. Identity verification and qualifying home-market spend may be required for international targeting.

What revenue has OpenAI disclosed?

Reuters reported OpenAI said its advertising business reached a $1 billion annualized revenue run rate in August. That is a run-rate estimate, not revenue earned over a full year or a result attributable to the October expansion.

References